Satellite Baseband Multipath Suppression via Deep Learning Image Mode
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current GNSS systems face challenges in accurately suppressing multipath signals in complex urban environments, leading to reduced positioning accuracy and limited adaptability across different scenarios.
Innovation Solution
A suppression method for multipath signals based on correlation peaks of a satellite baseband signal, utilizing a deep learning network with an LSTM network and self-attention mechanism, to construct a multipath suppression model that extracts time series and spatial aggregation features from a two-dimensional color heatmap, and fuses these features for effective multipath signal suppression.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional convolutional neural networks (CNN, AlexNet, VGG) are used for multipath identification, then the model can process signal data, but the performance for multipath identification task is not ideal and the trained model lacks reliability and robustness
Solution Approach 1:
The patent transforms the input data representation from traditional signal parameters to image mode data, changing the parameter space in which the neural network operates. This transformation enables the use of more effective deep learning architectures (CNN, AlexNet, VGG) that are optimized for image processing, thereby improving multipath identification accuracy and model reliability
Solution Approach 2:
The patent replaces traditional signal processing methods with deep learning-based image processing methods. By substituting the mechanical signal analysis approach with an intelligent image recognition system, the patent achieves superior multipath identification performance and model robustness
2Adaptability or versatility
If signal observation value is used as input for model training, then the model can be trained with available data, but the number of indicators is limited and it is difficult to directly and comprehensively reflect the signal interference situation
Solution Approach 1:
The patent adds a new dimension to the data representation by transforming one-dimensional signal observation values into two-dimensional image mode data. This dimensional transformation enables the model to capture spatial relationships and patterns that are not visible in traditional signal parameters, thereby comprehensively reflecting signal interference situations while maintaining adaptability to different scenes
3Reliability
If hardware-based methods (anti-multipath antenna or improving hardware performance) are used, then the suppression capability may be enhanced, but the application scenario is limited due to the requirement of hardware modification
Solution Approach 1:
The patent replaces hardware-based mechanical suppression methods with software-based deep learning methods. This substitution eliminates the need for hardware modifications while achieving effective multipath suppression, thereby maintaining suppression capability while significantly improving application scenario flexibility and adaptability
4Reliability
If software-based methods (sidereal filtering or multipath semi-sky sphere mapping) are used, then multipath suppression can be achieved, but the methods have problems such as instability or huge amount of calculation
Solution Approach 1:
The patent performs preliminary transformation of signal data into image mode representation before processing. This preliminary action organizes the data in a structured format that enables efficient processing by deep learning models, thereby improving calculation efficiency while maintaining suppression stability
Solution Approach 2:
The patent replaces traditional iterative software-based suppression methods with a deep learning-based approach. This substitution replaces complex iterative calculations with a trained model that provides stable and efficient suppression results, simultaneously improving both stability and productivity
Data Source
AI summary
Disclosed is a suppression method for a multipath signal of an image mode based on correlation peaks of a satellite baseband signal, including the steps of: constructing a real-world direct-multipath two-dimensional color image mode data set; and building a multipath suppression model of a deep learning network based on a long-short term memory (LSTM) and a self-attention mechanism module, and training the model. In the present disclosure, a satellite signal can be quickly captured without losing the sensitivity of the captured signal, and a Beidou satellite baseband signal can be captured from a signal recorded in the real world, which enables the model to learn complex signal patterns, improving the accuracy and robustness of multipath signal suppression in urban complex scenes.